AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Learning Differentiable Hierarchies in 3D Gaussian Splatting

Youqi Pan, Wugen Zhou, Hongbin Zha

CVPR 2026 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add hierarchy balance loss (ablation)DDSF hierarchy learning · Truck sequence (avg over splat ratios)PSNR: 21.76 -> 25.24Improved
use PDF-guided active-region sampling instead of random sampling (ablation)DDSF hierarchy learning · Truck sequence (avg over splat ratios)PSNR: 24.65 -> 25.24Improved
use PDF-guided sampling instead of quantile sampling (ablation)DDSF hierarchy learning · Truck sequence (avg over splat ratios)PSNR: 24.23 -> 25.24Improved
set sigmoid slope beta to 0.1 instead of 10 (ablation)DDSF · Truck sequence (avg over splat ratios)PSNR: 25.24 -> 22.49Hurt
set sigmoid slope beta to 1.0 instead of 10 (ablation)DDSF · Truck sequence (avg over splat ratios)PSNR: 25.24 -> 24.86Hurt
set sigmoid slope beta to 100 instead of 10 (ablation)DDSF · Truck sequence (avg over splat ratios)PSNR: 25.24 -> 24.11Hurt
use single-stage learned hierarchy instead of CLOD two-stage trainingLoD rendering · Mip-NeRF360, 100% splatsPSNR: 29.59 -> 29.74Improved
use single-stage learned hierarchy instead of CLODLoD rendering · Mip-NeRF360Train time (min:s): 59:34 -> 26:41Improved
add hierarchy learning to 3DGS-MCMC3DGS-MCMC · Mip-NeRF360Train time (min:s): 23:53 -> 26:41Hurt
use dedicated pruning training instead of 60% LoD model (ablation)Ours on 3DGS-MCMC · Mip-NeRF360 pruningPSNR: 29.59 -> 29.67Improved
use learned hierarchy instead of CLOD at 50% splats on Mip-NeRF360LoD rendering · Mip-NeRF360, 50% splatsPSNR: 29.51 -> 29.47No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy